I have no eyes and I must see!

February 7, 2010 10:36 PM

Sea urchins do not have eyes, yet appear to be able to see where they are going. One posible answer: they may use the entire surface of their bodies as a compound eye.
posted by Artw (31 comments total)

More sea urchin coverage from the echinoblog.
posted by Artw at 10:38 PM on February 7, 2010

"In the beginning, people built robots like they would humans, with powerful central processing units, complex sensors and fairly complex rules for doing things," Johnsen said. "Now they're finding it might be a lot better with a distributed system with many little processors and simpler sensors and simple rules, which end up creating fairly complicated behaviors as emergent properties"

Well, I guess maybe they designed them originally the way that humans THINK that they work. People actually are mostly a bunch of small, somewhat independent processors and sensors that create complicated behavior as a result of emergent properties.
posted by empath at 10:44 PM on February 7, 2010 [10 favorites]

This is probably the most shockingly amazing thing I have ever heard.
posted by stoneweaver at 10:48 PM on February 7, 2010

Robots should definitely be covered in spikes though.
posted by Artw at 10:48 PM on February 7, 2010 [1 favorite]

eww
posted by Crane Shot at 10:51 PM on February 7, 2010

Street urchins, on the other hand, do have eyes: Big, saucer-like eyes that stare from beneath their ragamuffin caps, before dashing off into a lively song-and-dance number involving fruit-sellers and chimney-sweeps. You were saying?
posted by bicyclefish at 11:34 PM on February 7, 2010 [15 favorites]

This is pretty cool. I love how half "raced" toward the spot ("Look! A Thing!") and half "raced" away ("Shit! A Thing!").
posted by exlotuseater at 11:41 PM on February 7, 2010 [11 favorites]

optics is so rad...i wonder if the fine structure on the spines (ridges, stripes, etc) helps with image quality (constructive diffraction? polarization?) and if this is why insects have those little hairs between the cells of their compound eyes.
posted by sexyrobot at 11:43 PM on February 7, 2010

This is really fascinating. Thank you for sharing!

How long before this information is exploited to make the ultimate urchin strobe lure by the Japanese?

Also: suck it, Creationists! *

* Some Creationists use the complex eye (e.g. mammalian eyes) as an argument against the theorem of Evolution, arguing that:

• no transitional forms have been found with partial eyes.
• a partial eye would be useless. What good is only 2% of an eye? It would have been a disadvantage, not an advantage; and
• since eyes are hugely complicated, they could not arise by chance mutation.

Of course, even a tiny bit of light sensitivity can have potential survival advantages (e.g. finding hiding places, avoiding predators, etc.). What good is only 2% of an eye? It's at least 100% better than only 1% vision, and probably many times better than zero vision at all, when it comes to your ability to feed, mate, and otherwise survive.

Or just ask someone who can only see faint outlines and shadows if that's more useful than total darkness. (or better yet, just use your own brain for a change and figure it out for yourself)
posted by Davenhill at 1:32 AM on February 8, 2010 [2 favorites]

Online Etymology Dictionary:

urchin

late 13c., yrichon "hedgehog," from O.N.Fr. *irechon (cf. Picard irechon, Walloon ireson, Hainaut hirchon), from O.Fr. herichun "hedgehog" (Fr. hérisson), formed with dim. suffix -on from V.L. *hericionem, from L. ericius "hedgehog," from PIE base *gher- "to bristle" (cf. Gk. kheros "hedgehog;" see horror). Still used for "hedgehog" in non-standard speech in Cumbria, Yorkshire, Shropshire. Applied throughout 16c. to people whose appearance or behavior suggested hedgehogs, from hunchbacks (1520s) to goblins (1580s) to bad girls (c.1530); meaning "poorly or raggedly clothed youngster" emerged 1550s, but was not in frequent use until after c.1780. Sea urchin is recorded from 1591 (a 19c. Newfoundland name for them was whore's eggs).

posted by pracowity at 1:50 AM on February 8, 2010 [1 favorite]

Also: suck it, Creationists!

Dear God, hi, it's me again.

Um, I still don't have anything to pray for on my own behalf, so, uh, you know how it goes, can you just continue to, you know, not let sea urchins get it on with octopi?

Thanks, and amen.
posted by UbuRoivas at 2:45 AM on February 8, 2010 [1 favorite]

Wow, fascinating.

They mentioned that they can't test the urchin's vision with clipped spines (they die), but I wonder what would happen if you artificially increased the number of spines? Use toothpicks or whatever. Would they need to "know" about the spines to benefit - like, proprioception (and therefore adding more won't help)? Or would simply having more light-blocking specificity be enough?

Even more interesting, if artificial spines prove useful, would they have to learn about them? Like, no benefit at first, but over time they get better? If so, that'd be pretty incredible - proof of learning without a brain, in fact!

Oh, and Davenhill - sorry, but I think if creationists were logical, then they wouldn't be creationists. Natural selection really - assume a group of creationists, some logical, some not. All the logical ones eventually realize their illogical position and switch, so you're only left with the creationists incapable of following logic.

Now let's just hope they don't start reproducing a la idiocracy? (granted 603 and all, but still).
posted by Arandia at 2:59 AM on February 8, 2010

I wonder what would happen if you artificially increased the number of spines? Use toothpicks or whatever.

This is why mother and I decided you can't have pets.
posted by le morte de bea arthur at 3:45 AM on February 8, 2010 [9 favorites]

"In the beginning, people built robots like they would humans, with powerful central processing units, complex sensors and fairly complex rules for doing things,"

I glanced at that and thought it was the start of some kind of future Robot Bible.

I'd love to see some time lapse video of sea urchins reacting to things. I wonder if we can now use this fact to run learning experiments. Say, hiding food near one shape and nothing near another.
posted by lucidium at 3:59 AM on February 8, 2010

The influence of spines wouldn't be a diffraction-style effect, as the spines and the gaps between them are MUCH larger than the wavelength of light. But the spines would cast a pattern of shadows on the photo-sensitive surface of the animal's body. This type of sense wouldn't be anything like our sense of vision, but sea urchins don't need to read anyway.

Adding extra spines is actually a good idea, it's not obvious (to me, anyway) what would happen.

By itself, learning without a "brain" is something that has already been observed. For starters, many animals blur the brain/no brain line. They may have one large ganglion and many smaller ones. Hell, we're that way too, only our one large ganglion is WAY bigger. More directly, individual neurons (and even small components of neurons) adapt to repeated stimuli. For example, simple responses like startle reflexes can involve only one nerve cell, and simple animals can adapt their responses to repeated startles. However, studying visual learning in sea urchins would still be remarkable.
posted by Humanzee at 7:31 AM on February 8, 2010

They mentioned that they can't test the urchin's vision with clipped spines (they die)
Hmmm, what if they put paint or whatever on the spikes to keep the light out.

Also: this is awesome.
posted by shothotbot at 7:48 AM on February 8, 2010

That's how I do it.
posted by Lipstick Thespian at 7:54 AM on February 8, 2010

Ocean floor urchin is watching you pee in the water.
posted by fleetmouse at 7:57 AM on February 8, 2010

"In the beginning, people built robots like they would humans, with powerful central processing units, complex sensors and fairly complex rules for doing things," Johnsen said. "Now they're finding it might be a lot better with a distributed system with many little processors and simpler sensors and simple rules, which end up creating fairly complicated behaviors as emergent properties

Now its been years since I was really up on this stuff, but this makes all kinds of sense to me. One goal of this research has got to be moving us down the road to an intelligent artifact and I think, based on the evolutionary analogy, starting with sensory input and making the sensory processing the priority is the way to go rather then building a brain here and a motor there and an eye in a third place, hooking them all up and hoping for the best.
posted by shothotbot at 8:03 AM on February 8, 2010

Well, I guess maybe they designed them originally the way that humans THINK that they work.

It's probably also to do with it the scale of the available technology. A small video camera forty years ago was the size of a suitcase. It's only much more recently that we've begun to approach the scales of biological sensors.
posted by le morte de bea arthur at 8:19 AM on February 8, 2010

Although experiments that clip spines off urchins would seem to make sense if one wanted to directly test how they contributed to vision, "we can't just remove spines — if we do, they get very sick and die, and don't behave naturally at all," Johnsen explained.

Stupid question, but, couldn't they just cover up the spines with something lightweight and removable, like straws?
posted by odinsdream at 9:09 AM on February 8, 2010

In this theory The spines are supposed to block light, so covering them up wouldn't really do much.
posted by Artw at 9:18 AM on February 8, 2010

roll d20 against beholder
posted by mwhybark at 9:23 AM on February 8, 2010

delicious spiky eyeballs.
posted by emeiji at 10:36 AM on February 8, 2010

"no transitional forms have been found with partial eyes"

They do talk a lot of made-up crap, don't they? "No transitional forms have been found" translates as "I know nothing about the subject and I'm going to pull a 'fact' out of my ass." The planarian has an example of an intermediate form between surface photosensitive cells and an actual eye structure. Its photosensitive cells are arranged inside a concave surface ("eye cup"), which helps with directional perception. All you need is to grow a lens in front of it and you've got an eye as we understand it. A lens can start as a translucent coating which contains a trapped liquid. It really is one of the easier evolutionary puzzles to work out.
posted by George_Spiggott at 10:58 AM on February 8, 2010 [4 favorites]

Obscure Sri Lankan paedophile Arthur C Clarke touched on this as an idea in a 1975 book, Imperial Earth. One of the characters hypothesised that a space radio telescope could be made with thousands of urchin-like spines on an asteroid.

In retrospect it was kind of a jumbled mess - though I remember it fondly. It also predicted we'd have iPhones by 2276, IIRC.
posted by Sebmojo at 12:21 PM on February 8, 2010

Metafilter: Look! A thing!

Metafilter: Shit! A thing!
posted by nebulawindphone at 12:44 PM on February 8, 2010 [1 favorite]

Obscure Sri Lankan paedophile Arthur C Clarke
Wait, what?
posted by shothotbot at 12:50 PM on February 8, 2010

My theory is that urchins require the ability to see so that they know how to dress appropriately.

For those not getting the joke, decorator urchins are known for grabbing stuff from their environment and attaching it to their bodies as a sort of camouflage. Good in the wild, funny as hell in a reef tank where the things you offer him are tiny plastic top hats and the like.

Sadly, none of my photos of this are online right now.
posted by quin at 1:41 PM on February 8, 2010 [1 favorite]

Obscure Sri Lankan paedophile Arthur C Clarke

Shit-stir much? A tabloid piece named the man such, which led to investigations, which led to there being nothing found. Do you have better information than references to the accusations in the Sunday Mirror?

I didn't think so.
posted by hippybear at 1:56 PM on February 8, 2010 [2 favorites]

I find it interesting that it's possible sea urchins are best understood as having compound characteristics: they are tactile and respond to touch the way a porcupine might, but also are able to react to light/dark in a way that suggests they see in some sense. In other words, traditional divisions between senses breakdown when considering their behavior, which presents a kind of phenomenological "field" view of how we understand the entire sensual/perceptual ecology a given organism. It's curious too that, as I understand it, genetically they appear to be close to starfish, sea anemones, etc., but that none of those creatures is thought to be able to see.
posted by HP LaserJet P10006 at 9:53 PM on February 8, 2010